EE 565 - Computer Communication Networks - Winter 2008

Project Option I: Spectrum Allocation in Cognitive Radio Networks

In this project, we will investigate the problem of spectrum allocation in cognitive radio networks. Cognitive radios function by sensing under-utilization of spectra and then occupying the available resources. The problem consists of two primary challenges: i) cognitive radio users, or secondary users, are not allowed to interfere with primary users (e.g. TVs, etc.) and ii) secondary users must minimize interference with each other and efficiently use the available resources. As background reading, please read the following papers:

1. I. A. Akyildiz, W.-Y. Lee, M. C. Vuran, and S. Mohanty, NeXt Generation/Dynamic Spectrum Access/Cognitive Radio Wireless Networks: A Survey, Computer Networks, 50(13), Sept, 2006.
2. R. J. Berger, Open Spectrum: A Path to Ubiquitous Connectivity, ACM Queue 1(3), May, 2003.
3. C. Peng, H. Zheng, and B. Y. Zhao, Utilization and Fairness in Spectrum Assignment for Opportunistic Spectrum Access, Mobile Networks and Applications, 11(4), August, 2006.
4. G. Cheng, W. Liu, Y. Li, and W. Cheng, Joint On-Demand Routing and Spectrum Assignment in Cognitive Radio Networks, in Proc. 2007 IEEE Conference on Communications (ICC'07), Glasgow, Scotland, June 2007.

For each part of the project, you will be responsible for writing the necessary simulation code (reusing code from previous projects is recommended) and answer the questions asked of you. You may use any reasonable programming language. Prepare a detailed project write-up (ideally 8-12 pages, 15 pages max) containing the following:

• A brief summary of each part of the project,
• Any assumptions you had to make, including parameter values,
• A summary of your results and explanation of any figures, and
• A description of your implementation.

Create a .zip, .rar, or .tar archive including (1) your project write-up, (2) your source code, (3) a text file with instructions to run your code, and (4) a summary of project contributions by each group member. Submit this archive according to the online submission instructions.

Spectrum Allocation as Graph Coloring
Investigate the spectral allocation problem as a graph coloring problem using the approach of Peng, Zheng, and Zhao (Ref 3). The authors propose an algorithm for spectrum allocation and provide several metrics yielding various versions of the algorithm. The primary goal of the channel allocation problem is to choose an optimal assignment of channels to users to maximize a utility function U which is a function of the channel assignment and the rewards achieved by users as a result of the assignment. In addition, a valid channel assignment must provide channel access that does not cause interferance with primary users or among secondary users. For detailed models and notation, refer to Ref 3.

You are responsible for the following:

• Read the paper by Peng, Zheng, and Zhao in detail.
• Implement the graph coloring algorithm and generate a figure to illustrate the result of the algorithm for an example network. Differentiate between the primary and secondary users and indicate the protection areas around the primary users and the corresponding channels (in the same color as the graph edges) in your figure.
• Discuss the applicability of these algorithms in a real cognitive radio network. In what ways are they useful? Do they need to be modified? Are the assumptions realistic? How do the models in this work align with those of those in Refs 1 and 2?
• Implement the proposed algorithm using the corresponding metrics to recreate the simulation results in Figures 4 through 8 in the paper. Compare your simulation results to those in the paper and comment on any differences.
• Suggest an additional utilization metric U and repeat the corresponding simulations for your metric. Comment on any additional information that your metric reveals about the channel assignment problem.
• If you were to improve on the proposed algorithm, what additional information or metrics would you take into consideration? See if you can refine the heuristic algorithm to improve the performance (e.g. include an extension to the algorithm that refines the assignment, introduce additional network information, etc.).

Joint Routing and Spectrum Assignment
Investigate the interactions between routing and spectral allocation using the approach of Cheng, Liu, Li, and Cheng (Ref 4). The authors propose a metric based on delay and provide a model to describe the channel assignment process.

You are responsible for the following:

• Read the paper by Cheng, Liu, Li, and Cheng in detail.
• Implement the algorithm and generate a figure to illustrate the results for an example network.
• Discuss the applicability of these algorithms in a real cognitive radio network. In what ways are they useful? Do they need to be modified? Are the assumptions realistic? How do the assumptions in this work align with those of the background reading materials?
• Implement the proposed algorithms to recreate the simulation results in the paper. Compare your simulation results to those in the paper and comment on any differences.
• If you were to improve on the proposed algorithms, what additional information or metrics would you take into consideration? See if you can achieve any improvement over the performance of the algorithms in the paper.